Anatomy of a red tide bloom off the southwest coast of Florida

C.S. Yentsch a,b, B.E. Lapointe c,*, N. Poulton a, D.A. Phinney a

a Bigelow Laboratory for Ocean Sciences, West Boothbay Harbor, ME 04575, USAb Plankton Research & Instruments, 504 Bahama Street, Key West, FL 33040, USAc Harbor Branch Oceanographic Institute, Florida Atlantic University, 5600 US Highway One, Fort Pierce, FL 34946, USA

Harmful Algae 7 (2008) 817–826

A R T I C L E I N F O

Article history:

Received 1 August 2006

Accepted 18 April 2008

Keywords:

Ataxonomic analysis

Chromophoric dissolved organic matter

Early warning system

Eutrophication

Flow cytometry

Karenia brevis

Particle absorbance

Visible reflected light

A B S T R A C T

A massive outbreak of Karenia brevis that had been ongoing for several months along the southwestern

coast of Florida was sampled in early September 2005 off Sanibel Island to assess the utility of bio-optical

features and ataxonomic analysis (quantification of eukaryotic and cyanobacterial picoplankton) by flow

cytometry in monitoring red tide blooms. Sea-surface sampling followed aircraft visual location of

discolored water. Within the most concentrated area of the bloom, chlorophyll a values exceeded

500 mg l�1, and concentrations of nitrate (0.3 mM � 0.0) and ammonium (<0.2 mM) were depleted

compared to high concentrations of total dissolved nitrogen, total dissolved phosphorus, and soluble reactive

phosphorus (141 � 34 mM, 16.5 � 2.5 mM, and 6.44 � 0.57 mM, respectively). Low water clarity in the bloom

(Secchi depth transparency 0.3 m, Kd estimated at 4.83 m�1) was strongly influenced by attenuation from

dinoflagellates as well as chromophoric dissolved organic matter (CDOM). The fact that the K. brevis bloom

occurred in lower-salinity (30 psu), high-nutrient waters implicates riverine transport of land-based

nutrients as a source of nutrient supplies that fueled or sustained the bloom. Throughout ongoing efforts to

advance modeling and technological capabilities that presently lack reliable predictive capability, bio-optical

remote sensing via aerial flyovers along with in-water sensor data can continue to provide accurate coverage

of relatively large temporal and spatial features. Flow cytometry can provide conservative (because of some

cell lysis), rapid, near-real-time validation of bloom components. The concentration and position of the

organisms, along with water mass scalars, can also help to diagnose factors promoting K. brevis bloom

development and dispersion.

� 2008 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Harmful Algae

journa l homepage: www.e lsev ier .com/ locate /ha l

1. Introduction

The theme of this paper is borrowed from earlier work of Rytheret al. (1958). Both papers address problems of eutrophication inmarine coastal waters; both involve observations and measure-ments of massive microalgal blooms; both, in the Eltonion sense(Elton, 1927), describe phenomena wherein a bloom speciesthreatens the diversity in a natural ecosystem. A surprisingoutcome of the early Ryther et al. (1958) paper – that eutrophica-tion had stimulated blooms of microalgae that were unpalatablefor commercially important shellfish in coastal waters of New York– prompted the decision for engineering hydrographic changes(flushing strategies for the inlet) that set the stage for sensibleconservation and improved benthic faunal production (U.S. FederalWater Pollution Control Administration, 1966; Ryther and Dun-stan, 1971; Ryther and Officer, 1981). A key difference between the

* Corresponding author. Tel.: +1 772 465 2400x276; fax: +1 772 468 0757.

E-mail address: lapointe@hboi.edu (B.E. Lapointe).

1568-9883/$ – see front matter � 2008 Elsevier B.V. All rights reserved.

doi:10.1016/j.hal.2008.04.008

earlier study in New York waters and Florida ‘‘red tide’’ blooms ofthe harmful alga, Karenia brevis, is that the role of culturaleutrophication in stimulating these blooms remains a controver-sial issue (Hu et al., 2006; Brand and Compton, 2007; Lapointe andBedford, 2007; Shrope, 2008). This paper similarly is contributed inthe spirit of providing information to help facilitate initiatives toreduce the impacts of red tide blooms along the southwest coast ofFlorida.

The urbanized southwest coast of Florida has a long history oftoxic blooms of the naked dinoflagellate K. brevis (Davis) G. Hansenet Moestrup, or ‘‘red tides’’ (Gunther et al., 1947; Rounsefell andNelson, 1966; Brand and Compton, 2007). Brevetoxins produced bythis organism promote massive fish kills and, when aerosolized,adversely affect human respiration and asphyxiate domesticanimals (Lee et al., 1989). Widespread concerns are mountingalong the southwest coast of Florida, where the local economystrongly depends upon tourism and recreational fisheries (Ander-son et al., 2000; Larkin and Adams, 2007). Regional authoritieswish to determine the causes and the sources of red tides in anattempt to stem the problem and minimize impacts.

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826818

Two important early publications serve as guideposts for thispaper. First, Ketchum and Keen (1948) reported unusually highconcentrations of total phosphorus associated with red tideblooms in coastal waters off Sarasota, FL. Because theseconcentrations greatly exceeded the concentrations in surround-ing water masses, these workers postulated that ‘‘the excessivenutrient content may be the result of terriginous contamination orfertilization of the waters’’. Second, Slobodkin (1953) studied long-term records of red tide events (1840–1952) and noted that afterperiods of heavy rainfall, major red tide events occurred,supporting the hypothesis of land drainage as a major cause forcoastal outbreaks. Slobodkin (1953) specifically noted the impor-tance of a buoyant freshwater plume in his conclusion that ‘‘adiscrete mass of water, with a salinity lower than that of normalGulf of Mexico water, is a necessary pre-requisite for theoccurrence of red tide off the Florida Coast’’.

Our study is also contributed to assist coastal resourcemanagers in continuing to improve early warning systems for K.

brevis blooms. The concept of early warning systems is not new.Previous researchers proposed use of remote sensing (Lillesandand Kiefer, 1979; Yentsch, 1979; Yentsch and Phinney, 1989) andparticle counting techniques such as flow cytometric analysis(Yentsch et al., 1983; Yentsch, 1989; Veldhuis and Kraay, 2000) todetect and track harmful algal blooms (HABs). Early warningsystems focusing on toxin detection as well as cell numbers havebeen in use for decades in bivalve surveillance and aquaculture asprotective measures for seafood safety (Shumway, 1990). Morerecently, development and dissipation of blooms of some harmfulspecies have been tracked using real-time remote monitoringnetwork programs (e.g. Springer et al., 2005; Cullen and Roesler,2006); remote sensing techniques have been refined specifically totrack K. brevis blooms (Craig et al., 2006); and the imaging-flow-cytometer, Flow-CAM1, has been used to monitor K. brevis

abundance (Buskey and Hyatt, 2006). Although satellite tracking,real-time remote monitoring networks, and other sophisticatedinstrumentation continues to be developed and refined to detect,track, and forecast K. brevis blooms (e.g. Stæhr and Cullen, 2003;Stumpf et al., 2003; Tomlinson et al., 2004; Hu et al., 2005), thesetechniques often cannot be used during and shortly after majorstorms (Fisher et al., 2006). Moreover, sea-surface sampling effortsremain a critical component of early warning systems regardless ofweather conditions (Fisher et al., 2006).

The present study provides a recent example of application ofbiological optics and ataxonomic analysis with flow cytometry, inconcert with analysis of environmental conditions, to examine the‘‘anatomy’’ of a K. brevis bloom—that is, the bloom componentsmost responsible for optical attenuation of visible reflected light.

2. Materials and methods

Blooms of K. brevis had been ongoing along much of thesouthwestern coast of Florida since January 2005 when they werefirst detected by satellite imagery (Fisher et al., 2006). Theseblooms followed record rainfall and physical impacts on thewatersheds of Lake Okeechobee and southwest Florida in 2004 as aresult of the overlapping paths of hurricanes Charley, Frances, andJeanne (Mallin and Corbett, 2006; Lapointe and Bedford, 2007).Massive water releases from Lake Okeechobee through theCaloosahatchee River and into coastal waters of southwest Floridavia the Franklin Lock occurred in Fall of 2004 and throughout muchof 2005 (Lapointe and Bedford, 2007; South Florida WaterManagement District, DBHYDRO database). The severe K. brevis

bloom that developed off southwest Florida in 2005 led to awidespread hypoxic zone and mortality of benthic communities,fishes, sea turtles, birds, and manatees (Rothschild, 2005; Hu et al.,

2006). The red tides persisted until late 2005 when high winds andwater column mixing associated with Hurricane Wilma finallybegan to disperse the bloom.

In early September 2005, bloom concentrations of K. brevis werereported in nearshore waters off Fort Myers, FL. During thefollowing days, we employed aircraft and sea sampling formeasurements of biological optics and water chemistry bothinside and outside the central region of a bloom along Lee County’scoastline. Aerial photographs were taken between 10:00 h and12:00 h on 7 September from flights at 152.4 m and 304.8 maltitude (e.g. Fig. 1), using a Sony Cyber-Shot fitted with a Carl ZeissVario-Tessar 3� zoom lens. Six stations were selected for sea-surface sampling based on observations from the aircraft. Surfacewater samples were collected on 8 September between 09:00 hand 12:00 h at six stations, each approximately 500 m off the LeeCounty coastline between Bonita Springs and Captiva Island(Table 1 and Fig. 2).

2.1. Environmental conditions

Surface and near-bottom water temperature, salinity, dissolvedoxygen (DO), and pH were measured on 8 September using a YSIModel 650 MDS data logger coupled with a submersible Model 600QS probe (Yellow Springs, OH). A Secchi disk was used to measurewater transparency. Replicate water samples (n = 2) were col-lected, 0.25 m below the surface (hereafter designated as surface)and at near-bottom (1.0 m above bottom) using a Niskin watersampler, and held in darkness on ice during transport to thelaboratory and overnight until analysis. On the following day, sub-samples of the water samples were filtered and then refrigerated orfrozen according to measurement protocol. Sub-samples for flowcytometry were flown on ice to Bigelow Laboratory for OceanSciences in West Boothbay Harbor, ME, for analysis withinappropriate time intervals.

For nutrient analyses, 100-ml sub-samples (n = 2) were gentlyfiltered through 0.7 mm Whatman GF/F filters, using a 60 ml.syringe, into clean 150-ml high-density polyethylene bottles at theHarbor Branch Oceanographic Institution (HBOI) in Ft. Pierce, FL. Thesamples were frozen and analyzed within 28 days for ammonium,nitrate + nitrite, soluble reactive phosphorus (SRP), total dissolvednitrogen (TDN), and total dissolved phosphorus (TDP) at NutrientAnalytical Services, Chesapeake Biological Laboratory, Center forEnvironmental Science, University of Maryland, Solomons, MD(NAS-CBL). A Technicon Auto-Analyzer II was used for determina-tion of nitrate and SRP (detection limits, 0.01 mM and 0.02 mM,respectively), and a Technicon TRAACS 800 was used for analysis ofammonium (detection limit, 0.21 mM) following the techniques ofKeefe et al. (2004). Sub-samples were also analyzed for totaldissolved nitrogen and total dissolved phosphorus at NAS-CBL usinga Technicon Auto-Analyzer II with detection limits of 1.43 mM forTDN, and 0.03 mM for TDP (Keefe et al., 2004).

Particulate matter on the GF/F filters from the K. brevis bloomencountered at Station 4 off Sanibel Island were analyzed for d15N(n = 3) to gain insights about potential nitrogen sources supportingthe bloom (Heaton, 1986; Dawson et al., 2002). These filters weredried in a Fisher Scientific IsotempTM oven at 60 8C for 48 h at HBOI.Analyses were completed by Isotope Services Inc., Los Alamos, NM,with a Carlo-Erba N/A Elemental Analyzer and a VG Isomass massspectrometer using Dumas combustion. The standard used forstable nitrogen isotope analysis was N2 in air; d15N values (as %)were calculated as follows:

d15N ¼ R sample

R standard� 1

� �� 103;

where R = 15N/14N.

Fig. 1. Aerial photograph taken off the southern tip of Sanibel Island on 7 September 2005. Note the front formed from the tip of the island and extending west into the Gulf of

Mexico. The white patches along the front are floating dead fish. The yellow water west of the front was caused by red tide. The dark water to the east was the plume flowing

from San Carlos Bay. At the top of the image is the Sanibel Causeway. (For interpretation of the references to color in this figure legend, the reader is referred to the web version

of the article.)

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826 819

Attenuation of incident solar irradiance through the watercolumn (resulting from light absorption and scattering) arefunctions of the wavelength of light (l). We estimated thedownwelling diffuse attenuation coefficient, Kd, for photosynthe-tically active radiation (PAR; l = 400–700 nm), from Secchi depthtransparency (Zd; units (m)) as follows:

Kd ¼1:5

Zd:

2.2. Phytoplankton abundance: chlorophyll a and cell number

Specific volumes were filtered (Whatman GF/F filters, 0.7 mmpore size) for measurement of chlorophyll a and phaeophytin.Filters from the samples were stored in plastic Falcon 1002 Petridishes and held frozen until analysis. Filters were extracted in 85%acetone, and the chlorophyll a extract was measured in vitro usingthe fluorometric method of Yentsch and Menzel (1963). Themethod was calibrated using a standard for chlorophyll a obtainedfrom spinach (Sigma Chemical Corp.).

Allometric ataxonomy is useful in quantifying specific groups ofphototrophs, such as pico-, nano-, and microplankton, in naturalphytoplankton assemblages based on their pigment fluorescence(Li, 1997, 2002; Ciotti et al., 2002; Vaillancourt et al., 2004).Because of its relatively large size (cell diameter �35 mm; Brandand Compton, 2007), K. brevis causes a major shift in the overallallometry of natural phytoplankton assemblages along the south-west Florida shelf, which often have abundant picoplankton (Heilet al., 2007). To allometrically assess phytoplankton abundance(total, K. brevis, and picoplankton), unfiltered surface water fromeach station was maintained in the dark at 4 8C for 4 days prior toflow cytometric analysis (due to shipment of the sample). Sampleswere not chemically preserved in order to minimize K. brevis celldisruption. Sub-samples for quantification of K. brevis and otherabundant phototrophs were analyzed using a FACScan flow

cytometer (Beckton Dickinson, San Jose, CA) equipped with a15 mW 488 nm air-cooled argon-ion laser. Red (chlorophyll)fluorescence emissions were collected using a 650-nm long pass(LP) filter, and orange (phycoerythrin) fluorescence emissionswere collected using a 575-nm band pass (BP) filter. The thresholdwas set on red fluorescence to identify only chlorophyll-containingparticles (i.e. microalgae and detritus that contained chlorophyll orother degrading pigments). Readings were taken in logarithmicmode and analyzed using CELLQuest software (Becton Dickinson,San Jose, CA). The volume of sample analyzed was determinedgravimetrically. Different regions were selected to accommodatethe ‘‘signatures’’ of the different phytoplankton present (K. brevis,eukaryotic picoplankton, and cyanobacteria). Abundances weredetermined from the count obtained within a specified gatedregion and the total volume analyzed. The interpretation of flowcytometric ‘‘signatures’’ for ataxonomic groups was performedusing methods previously described (Olson et al., 1993, 1985),based on different fluorescent signatures and cell size interpretedfrom forward-angle light scatter. The presence of K. brevis wasconfirmed using microscopic observation.

To determine the size of the microalgal cells analyzed by flowcytometry, the forward-angle light scatter data (indicative of cellsize) were normalized to a linear scale, using particles of knownsize, using a calibration equation that was obtained with standardcalibration beads and cultures (units, mm) as internal standards:

size ¼ ð0:0091 � log FSCÞ þ 1:6518;

where FSC = forward scatter, i.e. the light scattered <108 as a cellpassed through the laser beam.

2.3. Particle absorption

Particulate absorption sample filters were analyzed on a HitachiU3010 dual beam spectrophotometer using a blank, wetted, GF/Ffilter as reference (Yentsch and Phinney, 1989). Raw optical

Fig. 2. Map of the study area showing sampling stations on 8 September 2005 at Bonita Springs (Station 1), Lover’s Key (Station 2), Lynhall’s (Station 3), South Sanibel (Station

4), Bowman’s (Station 5), and San Carlos Bay (Station 6). Insets show the relation of the Lee County study area to the Caloosahatchee River, Peace River, and Lake Okeechobee.

Table 1Station locations, environmental data, nutrient concentrations, d15N of Karenia brevis bloom (Station 4), chlorophyll a, and spectral absorption data

n Station #

Bonita Springs 1 Lover’s Key 2 Lynhall’s 3 South Sanibel 4 Bowman’s 5 San Carlos Bay 6

Latitude 2683300 2683840 2683200 2684260 2684400 2684600

Longitude �8188470 �8188750 �8189580 �8280400 �8281400 �8280200

Depth (m) 1 1.5 2 1.5 1.6 1.6 1.6

Secchi depth (m) 1 1.5 1.6 1.4 0.3 1.1 1.1

K (m�1) 1 0.97 0.91 1.04 4.83 1.32 1.32

Temperature (8C)

Surface 1 29.6 29.8 29.5 30.3 29.4 29.7

Bottom 1 29.8 30.0 29.9 29.8 29.4 29.5

Salinity (%)

Surface 1 31.9 30.0 23.2 30.1 32.1 14.5

Bottom 1 32.0 31.0 25.7 31.5 32.1 27.1

pH

Surface 1 7.0 7.9 8.0 8.3 8.1 8.2

Bottom 1 7.6 7.9 8.0 8.2 8.1 8.0

DO (mg l�1)

Surface 1 3.4 4.9 6.2 12.6 7.0 6.8

Bottom 1 0.7 3.6 6.7 6.6 6.7 6.2

Ammonium (mM) 2 <0.2 <0.2 <0.2 <0.2 0.3 � 0.1

Nitrate (mM) 2 0.4 � 0.0 0.5 � 0.1 0.3 � 0.0 0.3 � 0.0 2.4 � 0.0

TDN (mM) 2 29.0 � 0.0 33.7 � 1.1 141.3 � 34.7 23.0 � 0.1 48.4 � 0.2

SRP (mM) 2 0.23 � 0.01 0.68 � 0.01 6.44 � 0.57 0.25 � 0.00 1.34 � 0.03

TDP (mM) 2 0.92 � 0.01 1.48 � 0.23 16.51 � 2.47 1.02 � 0.08 2.08 � 0.08

d115N (%) 3 7.83 � 0.54

Chlorophyll a (mg 1�1)

Fluorometric 2 12.9 � 1.4 7.3 � 4.5 336.6 � 140.0 12.9 � 1.4 9.0 � 1.3

Spectrophotometric 1 11.9 8.4 8.5 543.2 17.4 12.9

Absorption, a (m�1)

Particles (ap) 1 0.158 0.142 0.136 6.175 0.344 0.215

Chlorophyll a (a�p) 1 0.013 0.017 0.016 0.011 0.020 0.017

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826820

Fig. 3. Flow cytometer cell counts for cyanobacteria (Synechococcus sp.),

picoeukaryotes, Karenia brevis, and total phytoplankton for Stations 1–6 sampled

on 8 September 2005.

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826 821

density (ODf) values between 350 nm and 750 nm (1.6 nmresolution) were corrected for pathlength amplification by thefilter (beta correction) by

ODs ¼ 0:355 ODf þ 0:514 OD2f ;

where ODs is the beta-corrected optical density in suspension. Theabsorption coefficient for particles, ap (m�1), was then calculatedfrom ODs by

apðm�1Þ ¼ 2:3 ODspr2

V

� �100;

where ODs is the beta-corrected optical density in suspension, pr2

is the area of particles on the filter, and V is the volume filtered inml.

Chromophoric dissolved organic matter (CDOM) absorptionwas measured in the dual beam spectrophotometer using 10 cmquartz cuvettes after filtration through a Gelman Sterivex 0.2 mmcartridge (D’Sa and Lohrenz, 1999). Similarly filtered nanopurewater was used in the reference cell. Optical density was measuredbetween 200 nm and 750 nm (1.6 nm resolution) and theabsorption coefficient for CDOM, ay (m�1), was calculated by

ayðm�1Þ ¼ 2:3 OD� 10;

where OD is the optical density using the 10 cm sample cuvettes.Specific absorption by phytoplankton, a*, is calculated by

normalizing cell absorption to chlorophyll a content. This is ameasure of the cell’s photosynthetic activity and varies amongspecies. Specific absorption coefficients were estimated forsamples collected from each station by

a�p ¼ap675 ðm�1Þ

extracted chl a ðmg l�1Þ;

where ap 675 is the value of the spectral particulate absorptioncoefficient at 675 nm and chl a is the concentration of chlorophyll a

calculated from extracting the filter after spectrophotometricanalysis.

3. Results

On 7 September, we located a K. brevis bloom near Point Ybeloff Sanibel Island by aerial reconnaissance (Fig. 1). The sky wasovercast, winds were light (10–15 km h�1) and a north-to-southcurrent was estimated at <1 knot. The bloom was observed onthe northern edge of a plume of water coming from San CarlosBay. The front between the plume and bloom is seen in thephotographs as white patches that were later identified as deadfish. The extent of the plume into the open Gulf of Mexico couldnot be determined, but was visually apparent for 3–4 km slowlycurving to the northwest. The albedo of the plume appeared tobe less than that of the bloom, the bloom was yellow/brown incolor, and was organized in long rows of dense color alternatingwith lighter color. The main portion of the plume was dark,almost black. The eastern edge of the bloom was somewhatscalloped, probably the result of active interchange across thefront.

While sampling the six stations along the Lee County coast(Fig. 2) from a small boat on 8 September, we encountered theamber-colored water suggestive of dinoflagellates, as well asmany dead fish, large and small, off Sanibel Island at Station 4.Onboard observers experienced throat and eye irritation char-acteristic of the aerosol irritants from these organisms. Summedup, in terms of the scalars measured while on station, themajor differences between this station and the others werethe supersaturated oxygen concentration, extremely high

chlorophyll a concentrations and decrease in water transparency(Secchi disk) as well as a visual change in watercolor (Table 1).

Temperature (29.4–30.3 8C) and salinity (14.5–32.1 psu) at thesix coastal stations were consistent with conditions reported in thelong-term study of Rounsefell and Dragovich (1966) for the regionand time of year (Table 1). Salinities were less than in offshore Gulfwaters (35–36 psu; Ketchum and Keen, 1948), especially withinthe bloom at Station 4 (salinity 30 psu), because of freshening fromhigh precipitation and land-based runoff associated with previousstorm events (Fisher et al., 2006; Lapointe and Bedford, 2007;Table 1). Station 4 was also characterized by elevated DO (surface,12.6 mg l�1) and reduced Secchi depth transparency (0.3 m) incomparison to the other stations (range 3.4–7.0 mg DO l�1, Secchidepth 1.1–1.6 m). Near-bottom DO indicated hypoxic conditions atStation 1, which was the station with the lowest surface pH (7.0).Water transparency was extremely low throughout the study area.The estimated Kd (PAR) for all stations except Station 4 ranged from0.91 m�1 to 1.32 m�1, and was 4.83 m�1 at Station 4 (Table 1).

Dissolved inorganic N and P were similar among the stationsexcept for much higher SRP (6.44 � 0.57 mM) at bloom Station 4,and higher nitrate (2.40 � 0.0 mM) at Station 6 (Table 1). All stationswere relatively high in total dissolved organic nutrients, withextremely high values at Station 4 (141 � 35 mM TDN,16.5 � 2.5 mM TDP). The d15N data of the K. brevis bloom at Station4 averaged +7.83 � 0.54%, a value similar to that found in macroalgaealong this coastline in 2004 (Lapointe and Bedford, 2007).

The nutrient and chlorophyll a data from the six stations areshown in Table 1. The central portion of the buoyant plume isassumed to be the position of lowest observed salinity at Station 4.High values of chlorophyll a, DO, total dissolved nitrogen andphosphorus occurred here in the confluence of the mixed outflowsof the Caloosahatchee and Peace rivers. This is consistent withwhat was observed from the aircraft.

Phytoplankton biomass, as chlorophyll a concentration, wasmoderate (7.3–17.4 mg l�1) in the study area except for very highconcentrations (337–543 mg l�1) at bloom Station 4 (Table 1). Thehigh chlorophyll concentration in surface waters of Station 4 wassupported by high cell numbers of cyanobacterial (e.g. Synecho-

coccus spp.) and eukaryotic picoplankton (maximum cell dimen-sion <3 mm) and K. brevis (Fig. 3). As expected, given its large size(Fig. 4), K. brevis dominated the water optics at Station 4, despiteaccounting for only �5% of total phytoplankton cells, and causedmore optical forward scattering than the eukaryotic and prokar-yotic picoplankton (Fig. 5). Microscopic examination of sub-samples, before and after flow cytometric analysis, revealed lysis ofsome K. brevis cells, which would have underestimated abundance.

Fig. 4. Particle size distributions of phytoplankton greater than 10 mm at Stations 1–

6 along the Lee County Coast. Only at Station 4 is a dramatic increase observed in

cell size and quantity. The shift in the cell distribution from smaller (10–20 mm)

cells to a much larger cell type (25–40 mm) correlates with the observations of the

K. brevis bloom in the surface waters.

Fig. 6. Particulate spectral absorption at Stations 1–6 sampled on 8 September 2005.

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826822

Spectral curves for particle absorption all showed an absorptionband in the near infrared region (NIR), due to chlorophyllabsorption at 675 nm (Fig. 6). As expected, the chlorophyll bandfrom the bloom (Station 4) was more pronounced at 675 nm, withno sign of accessory chromoproteins, yet there was a much sharperrate of increase in absorbance from 570 nm to 450 nm, undoubt-edly due to the conjugated–carotenoid complex of dinoflagellates

Fig. 5. Flow cytometric results for bloom Station 4. (A) Dot plot of forward angle

light scatter vs. relative red fluorescence (indicative of chlorophyll), showing

specific ataxonomic components of the particle population (K. brevis,

picoeukaryotes, and cyanobacteria Synechococcus sp.); (B) histogram plot of

frequency vs. forward angle light scattering, showing the wide phytoplankton

distribution at this station, i.e. cells of two size ranges—K. brevis (35 mm) and

cyanobacteria (1–2 mm).

(Yentsch, 1962). Particle absorption rapidly increased from 570 nmto 410 nm, likely related to the accessory pigments of K. brevis

(Bjørnland et al., 2003). A second rapid increase in particleabsorption occurred in the near-ultraviolet (UV; 350–400 nm)light region, suggesting the presence of mycosporine-like pig-ments (Bandaranayake, 1998; Evens et al., 2001).

CDOM was highest at Station 4, and CDOM action spectra (ay)revealed an exponential increase in absorbance from the NIR to UVwavelengths (Fig. 7), characteristic of humic and fulvic acids innatural waters (Kirk, 1994). The CDOM brown coloration increasedabsorption between 500 nm and 350 nm. Station 4 also had anabsorption feature centered around 300 nm that has beenobserved in algal exudates, and may have been caused fromabsorption by aromatic amino acids (Yentsch and Reichert, 1962).This feature generally is not observed in samples from open oceanwaters; it may represent a non-conservative feature of CDOM(Yentsch and Reichert, 1962) that influences the slope of theconservative fraction of CDOM into the visible region.

Water transparency was extremely low throughout the studyarea (Table 1). Kd at Stations 1–6 ranged from 0.9 m�1 to 1.3 m�1;at bloom Station 4, Kd was 4.8 m�1. To estimate the causes of highattenuation it was assumed that the fate of the radiation wasabsorption by particles and dissolved yellow organics (CDOM),which is nominal for coastal waters receiving considerableamounts of riverine discharge. The spectral characteristics of bothwere used to define their relative importance.

Comparison of spectral absorption coefficients for particles (ap)and CDOM (ay) indicated that CDOM dominated near-UVwavelengths at all six stations, but in the visible region (using450 nm as a reference) CDOM absorption varied (Fig. 8). Moreover,at bloom Station 4, particle absorption was mostly related tophytoplankton cells so that major algal pigments bands, such aschlorophyll and carotenoids at blue wavelengths and chlorophyllat red wavelengths, exceeded the specific absorption of CDOM.Specific absorption by chlorophyll a (a�p) at all six stations rangedfrom 0.011 m�1 to 0.017 m�1; the lowest value occurring at Station4 (Table 1). Mean a�p was 0.0157 + 0.003 m�1, similar to the valuereported by Gallegos and Bergstrom (2005) for the potentially toxic

Fig. 7. UV/vis absorption spectra for CDOM at Stations 1–6 sampled on 8 September

2005.

Fig. 8. Specific absorption spectra for particles (ap) and CDOM (ay) at Stations 1–6 sampled on 8 September 2005.

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826 823

dinoflagellate Prorocentrum minimum (Pavillard) Schiller, butlower than means reported for open ocean phytoplanktonassemblages (0.026 m�1; Yentsch and Phinney, 1989; Gallegosand Bergstrom, 2005).

4. Discussion

The K. brevis bloom characterized in this 2-day ‘‘snapshot’’analysis occurred in lower-salinity, high-nutrient, high-CDOMsurface waters. CDOM along the western Florida shelf is mostlycontributed by riverine/estuarine sources (Del Castillo et al., 2000),implicating riverine transport of land-based nutrients as a majorsource of nutrition to the bloom. The shallow waters of the bloomarea had a lower salinity surface layer and a more saline underlyinglayer, promoting water column stability. Such conditions havebeen observed in many coastal regions experiencing culturaleutrophication (e.g. Ochi, 1989; Wyatt, 1990), and were identifiedby Slobodkin (1953) as a necessary pre-requisite for Florida redtides. Nutrients in CDOM from land drainage have also beenreported to favor some blooms (Anderson and Corbett, 1979). Forexample, in Swedish coastal waters, Graneli et al. (1989) found thatorganic N and P were major forms of nutrients in ‘‘brown water’’.Addition of CDOM favored growth of the dinoflagellate, P.

minimum, possibly due to CDOM organic nutrient and heavymetal chelator content (Heil, 2005). Ingle and Martin (1971)proposed an iron/CDOM index to predict Florida K. brevis bloomsunder conditions of iron limitation.

We earlier suggested a potential commonality between themicroalgal blooms observed by Ryther et al. (1958) in eutrophicwaters of Long Island (enriched by duck farm runoff) and the K.

brevis bloom sampled in this study off Sanibel Island on the westcoast of Florida. The red tide we sampled in lower-salinity (30 psu)waters had very high concentrations of TDP similar to valuesreported over a half-century earlier by Ketchum and Keen (1948).These authors found values of 14.6–20.4 mM TP in �32 psu waterof ‘‘deep amber color’’ associated with red tides some 1.5 miles offSarasota Point, Sarasota, FL. Those values bracket the mean value(16.5 � 2.47 mM) we observed for the red tide in similar ambercolored water off Sanibel Island in 2005. Considering that thesestudies were almost 60 years apart, the similarity in values confirmsthe high quality of P analyses obtained by oceanographers in thatperiod. The SRP concentrations associated with red tide off Sanibel in2005 were also very high, averaging 6.44 � 0.57 mM. That is muchhigher than SRP concentrations in the range of 0.30–0.92 mMreported for coastal waters of Lee County in 2004 (Lapointe andBedford, 2007). The unusually high concentrations within the red tidebloom could be due, in part, to breakage of the naked and delicate K.

brevis cells during syringe filtration, which would release SRP and TDPfrom internal sources. Regardless, the high background SRP (up to1 mM) and TDP (5 mM) concentrations observed by Lapointe andBedford (2007) for nearshore coastal waters off Lee County in late2004 would be considered eutrophic and capable of supporting adense K. brevis bloom of >106 cells l�1 (Vargo et al., 2008).

Because of high background SRP and TDP concentrations in theLee County’s nearshore coastal waters, algal blooms are strongly N-limited (Lapointe and Bedford, 2007). Although there are a varietyof potentially important land-based nitrogen sources to coastalwaters, the mean d15N value (+7.83%) of the K. brevis bloom offSanibel closely matched the d15N values of +6–8% for macroalgalblooms collected on beaches and shallow reefs along this coastline

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826824

(Lapointe and Bedford, 2007). These red tide d15N values are alsosimilar to values reported for algal tissue at the Ortona andFranklin locks along the Caloosahatchee River, which can beinfluenced by both ammonium-rich Lake Okeechobee releases aswell as in-basin sources (Lapointe and Bedford, 2007; Fig. 2).Although various hypothetical sources of N supporting K. brevis

blooms along the west coast of Florida have been suggested –nitrogen fixation by Trichodesmium sp. (Lenes et al., 2001), andshelf-break upwelling (Stumpf et al., 2008) – the d15N data formacroalgae in Lapointe and Bedford (2007), combined with thesimilar d15N data for K. brevis in this study, indicate that land-basedN sources are major contributors to both of these HABs innearshore waters. While far-field sources of nutrients mightcontribute to blooms in offshore waters, we suggest that thecombined d15N data for macroalgae and red tides in nearshorewaters represent a strong signal of cultural eutrophication(Heaton, 1986; Dawson et al., 2002) that may be linked to theincreasingly intense K. brevis blooms (Brand and Compton, 2007).These d15N values are enriched well above values for N fixation,fertilizer N, and atmospheric deposition but are within the range ofvalues reported for wastewater N from septic tanks, secondarilytreated effluent, and cattle farms (Lapointe and Bedford, 2007). Ourdata and interpretations are not new to the general problem ofcoastal eutrophication and Florida red tide literature. Most, if notall, of the early reports from the west coast of Florida that describeddead fish and discolored water associated the red tide phenomenawith rainfall and runoff into coastal waters (e.g. Ketchum and Keen,1948; Slobodkin, 1953). Our results support the analyses of Brandand Compton (2007) and Vargo et al. (2008) who also noted theimportance of land-based nutrient flux to the development ofdense K. brevis blooms in nearshore waters. That macroalgalblooms have also emerged in nearshore waters along the westcoast of Florida since 2003 (Lapointe and Bedford, 2007) supportsthe view that nutrient enrichment has recently reached (or,surpassed) a ‘‘tipping point’’ for high-biomass macroalgal blooms(Lapointe, 1997). Such blooms are known to reduce lighttransmission, promote bottom-water anoxia, and adversely effectbenthic fauna and biogeochemical cycles (Valiela et al., 1997;Lapointe and Bedford, 2007) and are considered harmful algalblooms, similar to red tides, even though the macroalgae lackdirect toxicity (ECOHAB, 1995).

Recent analyses of nutrient sources have provided otherinsights about the role of land-based nutrient pollution insupporting nearshore K. brevis blooms. Estimates of nutrientfluxes from groundwater (Hu et al., 2006) and estuarine surfacewaters (Vargo et al., 2004) support the premise that land-basednutrients may be significant in supporting nearshore K. brevis

blooms. Heil et al. (2001) reported that riverine inputs often wereinsufficient to support nearshore blooms; more recently, Vargoet al. (2008) estimated that the estuarine sources neverthelesscould be important in sustaining these blooms. By their calcula-tions, total nitrogen and total phosphorus fluxes from Tampa Bayand Charlotte Harbor could supply up to 20% of the N and up to�90% of the P needed to support a moderate K. brevis bloom(3 � 105 cells l�1, 0.2 divisions d�1). Hu et al. (2006) noted that thehurricanes of 2004 were followed by the unusually late onset andlong duration of red tides throughout 2005. The posthurricanedissolved inorganic nitrogen (DIN) inputs from Tampa Bay, alone,were estimated to have been �35% of the inputs from all Floridarivers draining west in combination. Hu et al. (2006) suggestedthat DIN carried into nearshore waters from the posthurricane-related high runoff, and higher-than-normal submarine ground-water discharge, helped to initiate and fuel the persistent K. brevis

blooms. Lapointe and Bedford (2007) also reported that theextensive red tides of 2005 followed large discharges from the

Caloosahatchee and Peace rivers following the 2004 and 2005hurricanes, which resulted in ammonium and SRP enrichment toconsiderable distances (>26 km) from shore.

We earlier stated our hope that this study of a K. brevis bloomcan provide direction to assist coastal resource managers inreducing red tide outbreaks. Ryther et al. (1958) suggested, foreutrophic Great South Bay in New York, that shellfish production(inhibited by blooms) would be improved if flushing wasincreased. As a result, an inlet was opened, shellfish productionincreased, and the resource protected. Thus, benthic resourceswere left intact and, in fact, were improved without need of othermanagement intervention potentially harmful to beneficial fauna(e.g. Lewis et al., 2003), or which might exacerbate bloomconditions by causing cell lysis and brevetoxin release, such asapplication of clay (e.g. Pierce et al., 2004; Sengco and Anderson,2004) or heavy metal toxicants like copper sulfate (Rounsefell andDragovich, 1966). Somewhat analogously, Slobodkin (1953)proposed a plan for estuaries along the west coast of Florida thatinvolved reducing red tides by altering estuarine flow velocitiesand flushing rates. He stressed that accurate prediction of K. brevis

blooms would require a thorough knowledge of estuarine inputs tothe affected coastal waters. Because of the established linkagesbetween land-based nutrient inputs and HABs off southwestFlorida, improved management of the quality and quantity of thefreshwater discharges from the watersheds, particularly theCaloosahatchee River, could help moderate blooms in the future.

The present study of a K. brevis bloom was also contributed withthe hope of assisting coastal resource managers in efforts tocontinue to improve early warning systems for K. brevis blooms.The simple combination of bio-optical remote sensing via aerialflyovers, together with in-water sensors and flow cytometry, asused in this study, can provide critically needed information on K.

brevis blooms during weather conditions that preclude reliance onsatellite imagery and other techniques. We have suggested aminimum set of measurements that can capitalize on thesignificant indicators of the presence of K. brevis and thehydrographic conditions under which it blooms in southwestFlorida. For cell size and concentrations, flow cytometric measure-ments of cell size distributions can be used to monitor populationsof the large K. brevis cells. For the presence of increasing amounts ofdinoflagellates, spectral absorption of the particulate and dissolvedfractions of seawater can indicate the presence of cellular andreleased mycosporine pigments. A decrease in the specificabsorption coefficient of the population (a�670) can also indicatethe dominance of dinoflagellates compared to picoplankton.Temperature, salinity, and nutrient concentrations of the watercolumn, combined with d15N signatures of K. brevis and various Nsources within a region, can be used to determine the presence andsource(s) of excess nitrogen loads that stimulate the blooms. Allthe sampling can be accomplished from small boats under weatherconditions that limit the efficacy of satellite imagery to forecastbloom components. Sample turnaround times are short such thatearly warning and monitoring capabilities are possible for thespecific application to K. brevis blooms in southwest Florida coastalwaters. Similar collections of minimum datasets for specificindicators will be needed for other bloom species in other coastalregions.

Acknowledgments

Funding for this research was provided by Lee County, FL. Thisstudy would not have been possible without the actions of LeeCounty officials, including County Commissioner Ray Judah; SmartGrowth Director Wayne Daltry; Director of the Division of NaturalResources, Roland Ottolini; and public opinion by way of the local

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826 825

press (Naples News-Press, Ft. Myers Press-Journal). Mr. KeithKibbey, laboratory manager of the Lee County EnvironmentalLaboratory, provided boat and analytical support as well aslaboratory space for this work. Flow cytometry analysis andassistance were provided by the J.J. MacIsaac Aquatic CytometryFacility at the Bigelow Laboratory for Ocean Sciences. Thismanuscript was improved by the comments of three anonymousreviewers and Mr. Brad Bedford. This is contribution number 1708of the Harbor Branch Oceanographic Institute at Florida AtlanticUniversity, Ft. Pierce, FL.[SS]

Addendum: Real-time data of this region are now available athttp://recon.sccf.org. The reader is directed to the results from anew partnership of citizens, Sanibel Captiva Conservation Founda-tion, government and business efforts. The River Estuary andCoastal Observation Network (RECON) is a network of opticalwater quality sensors (Satlantic, Inc.) deployed throughout theCaloosahatchee River and estuary to provide real-time waterquality data to scientists, policy makers and the general public.RECON’s network of high quality, autonomous, in situ sensors candetect the presence of algal blooms and nutrient hotspots. Anairborne companion to this network of activity has been proposedand is under consideration.

References

Anderson, D.M., Hoagland, P., Kaoru, Y., White, A.W., 2000. Estimated annualeconomic impacts from harmful algal blooms (HABs) in the United States.Technical Report WHOI-2000-11. Woods Hole Oceanographic Institute, WoodsHole, MA.

Anderson, D.M., Corbett, M.D., 1979. Workshop: the role of chelators and tracemetals in toxic blooms. In: Taylor, D.L., Seliger, H.H. (Eds.), Toxic DinoflagellateBlooms. Elsevier, New York, pp. 463–467.

Bandaranayake, W.M., 1998. Mycosporines: are they nature’s sunscreens? Nat.Prod. Rep. 15, 159–172.

Bjørnland, T., Haxo, F.T., Liaaen-Jensen, S., 2003. Carotenoids of the Florida red tidedinoflagellate Karenia brevis. Biochem. Syst. Ecol. 31, 1147–1162.

Brand, L., Compton, A., 2007. Long-term increase in Karenia brevis abundance alongthe southwest Florida Coast. Harmful Algae 7, 232–252.

Buskey, E.J., Hyatt, C.J., 2006. Use of the FlowCAM for semi-automated recognitionand enumeration of red tide cells (Karenia brevis) in natural plankton samples.Harmful Algae 5, 685–692.

Ciotti, A.M., Lewis, M.R., Cullen, J.J., 2002. Assessment of the relationships betweendominant cell size in natural phytoplankton communities and the spectralshape of the absorption coefficient. Limnol. Oceanogr. 47, 404–417.

Craig, S.E., Lohrenz, S.E., Lee, Z.P., Mahoney, K.L., Kirkpatrick, G.J., Schofield, O.M.,Seward, R.G., 2006. Use of hyperspectral remote sensing reflectance for detec-tion and assessment of the harmful alga, Karenia brevis. Appl. Opt. 45, 5414–5425.

Cullen, J.J., Roesler, C.S. (Eds.), 2006. Real-time Coastal Observing Systems forEcosystem Dynamics and Harmful Algal Blooms. UNESCO Series—Monographson Oceanographic Methodology. Intergovernmental Oceanographic Commis-sion of UNESCO, Paris, France.

Dawson, T.E., Mambelli, S., Plamboeck, A.H., Templer, P.H., Tu, K.P., 2002. Stableisotopes in plant ecology. Annu. Rev. Ecol. Syst. 33, 507–559.

Del Castillo, C.E., Gilbes, F., Coble, P.G., Muller-Karger, F.E., 2000. On the dispersal ofriverine colored dissolved organic matter over the west Florida shelf. Limnol.Oceanogr. 45, 1425–1432.

D’Sa, E.J., Lohrenz, S.E., 1999. Lasers and laser optics—theoretical treatment offluorescence detection by a dual-fiber-optic sensor with consideration ofsampling variability and package effects associated with particles. Appl. Opt.38 (12), 2524.

ECOHAB, 1995. The ecology and oceanography of harmful algae blooms. In:Anderson, D.M. (Ed.), A National Research Agenda. WHOI, Woods Hole, MA.

Elton, C., 1927. Animal Ecology. Sidgwick and Jackson Publishers, London.Evens, T.J., Kirkpatrick, G.J., Millie, D.F., Chapman, D.J., Schofield, O.M.E., 2001.

Photophysiological responses of the toxic red-tide dinoflagellate Gymnodiniumbreve (Dinophyceae) under natural sunlight. J. Plankton Res. 23, 1177–1193.

Fisher, K.M., Allen, A.L., Keller, H.M., Bronder, Z.E., Fenstermacher, L.E., Vincent, M.S.,2006. Annual Report of the Gulf of Mexico Harmful Algal Bloom OperationalForecast System (GOM HAB-OFS). October 1, 2004 to September 30, 2005(Operational Year #1). NOAA Technical Report NOS CO-OPS 047. NOAA, SilverSpring, MD.

Gallegos, C.L., Bergstrom, P.W., 2005. Effects of a Prorocentrum minimum bloom onlight availability for and potential impacts on submersed aquatic vegetation inupper Chesapeake Bay. Harmful Algae 4, 553–574.

Graneli, E., Ollsson, P., Sunstrom, B., Edler, L., 1989. In situ studies of the effects ofhumic acids on dinoflagellates and diatoms, in red tides, biology. In: Okaichi,

T., Anderson, D.M., Nemoto, T. (Eds.), Environmental Science and Toxicology.Elsevier, pp. 209–212.

Gunther, G., Williams, R.H., Davis, C.C., Smith, F.G.W., 1947. Catastrophic massmortality of marine animals and coincident phytoplankton bloom on the westcoast of Florida. Ecol. Monogr. 18, 311–324.

Heaton, T.H.E., 1986. Isotopic studies of nitrogen pollution in the hydrosphere andatmosphere: a review. Chem. Geol. 59, 87–102.

Heil, C.A., 2005. Influence of humic, fulvic and hydrophilic acids on the growth,photosynthesis and respiration of the dinoflagellate Prorocentrum minimum(Pavillard) Schiller. Harmful Algae 4, 603–618.

Heil, C.A., Revilla, M., Glibert, P.M., Murasko, S., 2007. Nutrient quality drivesdifferential phytoplankton community composition on the southwest Floridashelf. Limnol. Oceanogr. 53, 1067–1078.

Heil, C.A., Vargo, G.A., Spence, D., Neely, M.B., Merkt, R., Lester, K., Walsh, J.J., 2001.Nutrient stoichiometry of a Gymnodinium breve Davis (Gymnodiniales: Dino-phyceae) bloom: what limits blooms in oligotrophic environments? In: Halle-graeff, G., Blackburn, S.I., Bolch, C.J., Lewis, R.J. (Eds.), Harmful Algal Blooms2000. Intergovernmental Oceanographic Commission of UNESCO 2001, Paris,pp. 165–168.

Hu, C., Muller-Karger, F.E., Swarzenski, P.W., 2006. Hurricanes, submarine ground-water discharge, and Florida’s red tides. Geophys. Res. Lett. 33, L11601,doi:10.1029/2005GL025449.

Hu, C., Muller-Karger, F.E., Taylor, C., Carder, K.L., Kelble, C., Johns, E., Heil, C.A.,2005. Red tide detection and tracing using MODIS fluorescence data: aregional example in SW Florida coastal waters. Remote Sens. Environ. 97,311–321.

Ingle, R.M., Martin, D.F., 1971. Prediction of the Florida red ride by means of the ironindex. Environ. Lett. 1, 69–74.

Keefe, C.W., Blodnikar, K.L., Boynton, W.R., Clark, C.A., Frank, J.M., Kaumeyer, N.L.,Weir, M.M., Wood, K.V., Zimmermann, C.F., 2004. Nutrient analytical serviceslaboratory standard operating procedures. Special Publication Series SS-80-04-CBL. Chesapeake Biological Laboratory, Center for Environmental and EstuarineStudies, The University of Maryland System, Solomons, MD.

Ketchum, B.H., Keen, J., 1948. Unusual phosphorus concentrations in the Florida‘‘red tide’’ seawater. J. Mar. Res. 7, 17–21.

Kirk, J.T.O., 1994. Light and Photosynthesis in Aquatic Ecosystems, 2nd ed. Cam-bridge University Press, New York.

Lapointe, B.E., 1997. Nutrient thresholds for bottom-up control of macroalgalblooms on coral reefs in Jamaica and southeast Florida. Limnol. Oceanogr. 42(5 part (2)), 1119–1131.

Lapointe, B.E., Bedford, B.J., 2007. Drift rhodophyte blooms emerge in Lee County,FL, USA: evidence of escalating coastal eutrophication. Harmful Algae 6,421–437.

Larkin, S.L., Adams, C.M., 2007. Harmful algal blooms and coastal business: eco-nomic consequences in Florida. Soc. Nat. Resour. 20 (9), 849–859.

Lee, M.S., Qin, G., Nakanishi, K., Zagorski, M.G., 1989. Biosynthesis studies ofbrevetoxins, potent neurotoxins produced by the dinoflagellate Gymnodiniumbreve. J. Am. Chem. Soc. 111, 6234–6241.

Lenes, J.M., Darrow, B.P., Cattrall, C., Heil, C.A., Callahan, M., Vargo, G.A., Byrne, R.H.,Prospero, J.M., Bates, D.E., Fanning, K.A., Walsh, J.J., 2001. Iron fertilization andthe Trichodesmium response on the West Florida shelf. Limnol. Oceanogr. 46,1261–1277.

Lewis, M.A., Dantin, D.D., Walker, C.C., Kurtz, J.C., Greene, R.M., 2003. Toxicity of clayflocculation to the toxic dinoflagellate, Karenia brevis, to estuarine invertebratesand fish. Harmful Algae 2, 235–246.

Li, W.K.U., 1997. Cytometric diversity in marine ultraphytoplankton. Limnol. Ocea-nogr. 42, 874–880.

Li, W.K.U., 2002. Macroecological patterns of phytoplankton in the northwesternNorth Atlantic. Nature 419, 154–157.

Lillesand, T.M., Kiefer, R.W., 1979. Remote Sensing and Image Interpretation. JohnWiley and Sons, New York.

Mallin, M., Corbett, C.A., 2006. How hurricane attributes determine the extent ofenvironmental effects: multiple hurricanes and different coastal systems.Estuaries Coasts 29 (6A), 1046–1061.

Ochi, T., 1989. The development of anoxic water and red tide associated witheutrophication in Hiuchi Nada, The Seto Inland Sea, Japan. In: Okaichi, T., An-derson, D.M., Nemoto, T. (Eds.), Red Tides, Biology, Environmental Science andToxicology. Elsevier, New York, pp. 201–204.

Olson, R.J., Vaulot, D., Chisholm, S.W., 1985. Marine phytoplankton distributionsmeasured using shipboard flow cytometry. Deep-Sea Res. A 32, 1273–1280.

Olson, R.J., Zettler, E.R., DuRand, M.D., 1993. Phytoplankton analysis using flowcytometry. In: Kemp, P.F., Sherr, B.F., Sherr, E.B., Cole, J.J. (Eds.), Handbook ofMethods in Aquatic Microbial Ecology. Lewis Publishers, Boca Raton, FL, pp.175–186.

Pierce, R.H., Henry, M.S., Higham, C.J., Blum, P., Sengco, M.R., Anderson, D.M., 2004.Removal of harmful algal cells (Karenia brevis) and toxins from seawater cultureby clay flocculation. Harmful Algae 3, 141–148.

Rothschild, W., 2005. Red Tide Kills 79 Manatees of Southwest Florida in 2005.Sarasota Herald Tribune, Sarasota, FL.

Rounsefell, G.A., Dragovich, A., 1966. Correlation between oceanographic factorsand abundance of the Florida red-tide (Gymnodiuium breve Davis), 1954–61.Mar. Sci. 16, 402–413.

Rounsefell, G.A., Nelson, W.R., 1966. Red tide research summarized to 1964 includ-ing an annotated bibliography. US Fish Wildl. Ser. Spec. Sci. Rep. Fish 535.

C.S. Yentsch et al. / Harmful Algae 7 (2008) 817–826826

Ryther, J.H., Dunstan, W.M., 1971. Nitrogen, phosphorus, and eutrophication in thecoastal marine environment. Science 171, 1008–1012.

Ryther, J.H., Officer, C.B., 1981. Impact of nutrient enrichment on water uses. In:Neilson, B.J., Cronin, L.E. (Eds.), Estuaries and Nutrients. Humana Press, Clifton,NJ, USA, pp. 247–261.

Ryther, J.H., Yentsch, C.S., Hulburt, E.M., Vaccaro, R., 1958. The dynamics of a diatombloom. Biol. Bull. 115, 257–268.

Sengco, M.R., Anderson, D.A., 2004. Controlling harmful algal blooms through clayflocculation. J. Eukaryot. Microbiol. 51, 169–172.

Shrope, M., 2008. Red tide rising. Nature 452, 24–26.Shumway, S.E., 1990. A review of the effects of algal blooms on shellfish and

aquaculture. J. World Aquacult. Soc. 21, 65–104.Slobodkin, L.B., 1953. A possible initial condition for red tides on the coast of Florida.

J. Mar. Res. 12, 148–155.Springer, J.J., Burkholder, J.A.M., Glibert, P.M., Reed, R.E., 2005. Use of a real-time

remote monitoring network (RTRM) and shipborne sampling to characterize adinoflagellate bloom in the Neuse Estuary, North Carolina, USA. Harmful Algae4, 533–552.

Stæhr, P.A., Cullen, J.J., 2003. Detection of Karenia mikimotoi by spectral absorptionsignatures. J. Plankton Res. 25 (10), 1237–1249.

Stumpf, R.P., Litaker, R.W., Lanerolle, L., Tester, P.A., 2008. Hydrodynamic accu-mulation of Karenia off the west coast of Florida. Continent. Shelf Res. 28,189–213.

Stumpf, R.P., Culver, M.E., Fester, P.A., Tomlinson, M., Kirkpatrick, G.J., Pederson, B.A.,Truby, E., Ransibrahmanakul, V., Soracco, M., 2003. Monitoring Karenia brevisblooms in the Gulf of Mexico using satellite ocean color imagery and other data.Harmful Algae 2, 147–160.

Tomlinson, M.C., Stumpf, R.P., Ransibrahmanakul, V., Truby, E.W., Kirkpatrick, G.J.,Pederson, B.A., Vargo, G.A., Heil, C.A., 2004. Evaluation of the use of SeaWiFSimagery for detecting Karenia brevis harmful algal blooms in the eastern Gulf ofMexico. Remote Sens. Environ. 91, 293–303.

U.S. Federal Water Pollution Control Administration, 1966. Report on pollution ofthe navigable waters of moriches bay and eastern section of Great South Bay,Long Island, New York. Hudson-Champlain and Metropolitan Coastal Compre-hensive Water Pollution Control Project, Metuchen, NJ.

Vaillancourt, R.D., Brown, C.W., Guillard, R.R.L., Balch, W.M., 2004. Lightbackscattering properties of marine phytoplankton: relationships tocell size, chemical composition and taxonomy. J. Plankton Res. 26,191–212.

Valiela, I., McClelland, J., Hauxwell, J., Behr, P.J., Hersh, D., Foreman, K., 1997.Macroalgal blooms in shallow estuaries: controls and ecophysiological andecosystem consequences. Limnol. Oceanogr. 42, 1105–1118.

Vargo, G.A., Heil, C.A., Fanning, K.A., Dixon, L.K., Neely, M.B., Lester, K., Ault, D.,Murasko, S., Havens, J., Walsh, J., Bell, S., 2008. Nutrient availability in support ofKarenia brevis blooms on the central West Florida shelf: what keeps Kareniablooming? Continent. Shelf Res. 28, 73–98.

Vargo, G.A., Heil, C.A., Ault, D.N., Neely, M.B., Murasko, S., Havens, J., Lester, K.M.,Dixon, L.K., Merkt, R., Walsh, J., Weisberg, R., Steidinger, K.A., 2004. Four Kareniabrevis blooms: a comparative analysis. In: Harmful Algae 2002, Florida Fish andWildlife Conservation Commission, Florida Institute of Oceanography and IOCof UNESCO, pp. 14–16.

Veldhuis, M.J.W., Kraay, G.W., 2000. Application of flow cytometry in marinephytoplankton research: current applications and future perspectives. Sci.Mar. 64, 121–134.

Wyatt, 1990. Modelling the spread of red tides—a comment. In: Graneli, E.,Sundstrom, B., Edler, L., Anderson, D.M. (Eds.), Toxic Marine Phytoplankton.Elsevier, New York, pp. 253–256.

Yentsch, C.S., 1962. Measurement of visible light absorption by particulate matter inthe ocean. Limnol. Oceanogr. 7, 207–217.

Yentsch, C.S., 1979. Biochemical factors associated with remote sensing of colorchange and fluorescence. In: Taylor, D., Seliger, H. (Eds.), Toxic DinoflagellateBlooms. Elsevier, North Holland, pp. 297–302.

Yentsch, C.M., 1989. Detection, enumeration and quantification of cell properties byautomated analysis. In: Okaichi, T., Anderson, D.M., Nemoto, T. (Eds.), RedTides, Biology, Environmental Science, and Toxicology. Elsevier, New York,pp. 221–224.

Yentsch, C.M., Horan, P.K., Muirhead, K., Dortch, Q., Haugen, E.M., Legendre, L.,Murphy, L.S., Phinney, D., Pomponi, S.A., Spinrad, R.W., Wood, A.M., Yentsch,C.S., Zahurenec, B.J., 1983. Flow cytometry and sorting: a powerful techniquewith potential applications in aquatic sciences. Limnol. Oceanogr. 28, 1275–1280.

Yentsch, C.S., Menzel, D., 1963. A method for the determination of phytoplanktonchlorophyll and phaeophytin by fluorescence. Deep-Sea Res. 10, 221–231.

Yentsch, C.S., Phinney, D.A., 1989. A bridge between ocean optics and microbialecology. Limnol. Oceanogr. 34, 1698–1709.

Yentsch, C.S., Reichert, C., 1962. The interrelationship between water-solubleyellow substances and chloroplastic pigments in marine algae. Bot. Mar. 3,65–74.